Alibaba Cloud Memory Service Upgrade: Enhancing Long-Term Memory and Personalization for Personal Agents
Alibaba Cloud announced an upgrade to its Memory service, enhancing memory capabilities for user preferences, plans, and historical agreements in Personal Agent scenarios, and providing integration methods such as MCP and CLI. Memory extracts key information from conversations, saves it as memories and user profiles according to rules, and in subsequent sessions retrieves it based on user identity and the current question and adds it to the task context. In the HaluMem-medium comparison, Memory achieved the highest scores in all four metrics: recall, accuracy, and F1 for memory extraction, and the C metric for memory question answering; in the PersonaMem-v2 evaluation, it achieved an overall score of 56.82, ranking first among the 8 systems listed and highest in 7 of the 9 task dimensions. Developers can use the API to write, view, and semantically retrieve memories, and can also integrate with agent tools and frameworks such as Codex, DeepSeek Harness, Claude Code, and OpenClaw via MCP and CLI.
Why read this
AI score 51
The original article provides evaluation comparisons of memory extraction and profiling, as well as MCP and CLI integration methods, which can be used to assess how long-term memory is implemented in personal agents.
Topics
Body · AI translation
Add crowdsourced testing promotional copy Add crowdsourced testing promotional copy end Add App-exclusive traffic redirect
IT Home, November 2 news: Alibaba Cloud announced today that it has recently upgraded its Memory service: for Personal Agent scenarios, it enhances the ability to remember user preferences, plans, and historical agreements, and provides access methods such as MCP and CLI.
IT Home has learned that Memory extracts key information from conversations and saves it as memories and user profiles according to rules. In subsequent sessions, the Agent retrieves relevant memories based on the user's identity and the current question, and adds them to the task context to generate responses. New valid information can continue to be written into memory, helping the Agent continuously follow up on user needs and reducing repeated inquiries.
In recent HaluMem-medium comparison results, Memory achieved the highest scores in all four metrics: recall, accuracy, and F1 for memory extraction, as well as the memory question-answering C metric; in the PersonaMem-v2 evaluation, Memory ranked first among the 8 systems listed with an overall score of 56.82, and achieved the highest score in 7 of the 9 task dimensions.
Taking the financial advisory scenario as an example, when a user states "prefers stability, may need the money within half a year, and does not want too much fluctuation," Memory can save the risk preference, fund usage period, and liquidity needs as a user profile. In a later new session, the Agent can call the relevant profile and, combined with product materials and business rules, carry out comparisons and risk reminders. When the user's goals or preferences change, the Agent can first confirm and then update the memory. Among these, Memory is responsible for information storage and recall, while product comparison and risk reminders are completed by the Agent.
Developers can customize memory rules and profile fields, specify the information that needs to be saved, and view memory content, debug retrieval results, and track changes in the console, making long-term memory manageable and traceable.
In terms of access, developers can complete memory writing, viewing, and semantic retrieval through APIs, and can also integrate memory capabilities into Agent tools and frameworks such as Codex, DeepSeek Harness, Claude Code, and OpenClaw through MCP and CLI, integrating them into existing workflows.
Memory can also be used in conjunction with Skill: Memory saves the user's individual preferences, and Skill encapsulates reusable task processes. For example, developers can encapsulate steps such as reading user profiles, analyzing product risks, and checking requirement matching into a financial advisory Skill, calling the user's latest profile each time it is executed, so that service methods can be reused with tasks and service content can be adjusted with the user. Statement: This is an exclusive article of Sina.com. Reproduction without authorization is prohibited. tech_web_article_v2015_block_bottom Non-targeted 300*250 button 17/09 wenjing begin
Non-targeted 300*250 button end